update-installed-skills

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly states the tool checks upstream sources and will clone/update skills from GitHub/upstream repos (e.g., via npx ai-agent-skills check and sync), so the agent fetches and installs untrusted public repository content that can change behavior.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.80). Yes — the workflow explicitly states that skills installed from GitHub will be freshly cloned during sync (e.g., https://github.com/...), meaning remote repository code is fetched at runtime and can directly control agent behavior or execute code.